Characterisation of lupin anthracnose caused by Colletotrichum gloeosporioides

By

CHARCTERIZATION OF LUPIN ANTHRACNOSE CAUSlED BY

COLLETOTRICHUM GLOEOSPORIOIDES

DA WIT SOLOMON GHEBREMARIAM

Submitted in partial fulfillment of the requirements for the degree

MAGISTER SCIENTEAE AG~CUL TURAE

In the Faculty of Natural and Agricultural Sciences ..

Department of Plant Pathology University of the Free State Bloemfontein, South Africa

SUPERVISOR: PROF. W.J. SWART

CO -SUPERVISOR: DR. S. KOCH

Un1ver'ltelt van die

OronJe-Vrystaat

BlO:MFOt4TE

1

N

2 6 APR 2002

DECLARATION

"1 declare that the thesis hereby submitted by me for the degree Master of Science in Agriculture at the University of the Free State is my own independent work and has not

_..

previously been submitted by me at another University/Faculty. I further cede copyright of the thesis in favour of the University of the Free State."

6••••••••••••••• ••••••••••••••••••••••••••••••••

1.1 INTRODUCTION 2

1.2 THE INFECTION PROCESS ----_._----_ .. _---_ __ __ ..-_ --_ _-_ -_ .3

T

ABLE OF CONTENTS ACKNOWlLEJI)GEMENTS IV LiST OF T ABlLES v LiST OF FIGlURES VI GENERAL INTROJl)UCTION V111 REFERENCES x CHAPTER 1

A REVIEW OF THE INFECTION AND PENETRATION PROCESS IN THE GENUS COLLETOTRICHUM

1.2_1 Spore survival and dissemination ---)

1.2.2 Spore germination and appressorial formation 6

1_2.2_1 Spore germination 6

1_2_2_2 Spore adhesion 7

1.2_2_3 Appressorial formation 12

1_2_3A Germination and appressorium formation 15

1.3 PENETRATION AND POST PENETRATION REACTIONS 22

1_3.1 Mode of penetration 22

1_3_2 Mechanisms of penetration 22

1.3_3 Latency

25

1.3 A Infection strategies 29

1_3 A_I Biotrophic phase _31

1.4 RESISTANCE REACTIONS OF PLANTS TOWARDS

COLLETOTRlCHUM SPECiES

38

1.4.1 Hypersensitive reaction and phytoalexin accumulation .38 1.4.2 Epicatechin and Polygalacturonase Inhibiting Protein (PGIPL 40

1.4.3 Structural defense mechanisms 41

1.5 CONCLUSIONS

44

1.6 REFERENCES

..-_.__.---_ ..---_ ---_ ..---_ -_._-_ ..

---_.---_

---_ .._----_ .

47

CHAPTER2

PATHOLOGICAL CHARACTERIZATION 01\ THE SOUTH AFRICAN

POPULATION OF COLLETOTRICHUM GLOEOSPORIOIDES ASSOCIATED

WITH LUPIN ANTHRACNOSE TO SIXTEEN LUPIN CULTIV ARS

2.1

INTRODUCTION

76

2.2

MATERIALS AND METHODS

78

2.2.1 Collection of isolates and inoculum preparation 78

2.2.2 Soil preparation and planting 79

2.2.3 Inoculation 79

2.2.4 Disease assessment ---_ -_ ----_ __ __ ---- -_._---_ _----_ ..80

2.3

RESULTS

80

2.4

DISCUSSION

81

CHAPTER3

THE PRODUCTION OlFGIBBERELLIN-LIKE SU.BSTANCES BY

COLLETOTRlCHflM GLOEOSPOR/O/DES

ASSOCIATED WITH LUPIN

ANTHRACNOSE

3.1 INTRODUCTION

96

3.2 MATERIALS AND METHODS

97

3.2.1 Gibberellin extraction process , 98

3.2.2 a-Amylase assay 98

3.3 RESULTS

100

3.4 DISCUSSION

101

3.5 REFERENCES

...---_ ..---_ ..---_ ..---_ __ ---_ _----_ -_ ----_ --_ _-_.- 105

CHAPTER4

INFECTION PROCESS OF

C. GLOEOSPOR/O/DES

ASSOCIATED WITH

LUPIN ANTHRACNOSE ON TWO DIFFERENT

LUP/NUS

SPP. VARYING IN

SUSCEPTIBILITY

4.1 INTRODUCTION

117

4.2 MATERIALS AND METHODS

...--_ ----_ -_ _-_ -_..__ -- ---_ .118 4.2.1 Inoculum production and artificial inoculation of plants

J

18

4.2.2 Scanning electron microscopy

.1

19

4.2.2 Fluorescence microscopy

.1

19

4.3 RESULTS AND DISCUSSION

120

4.5 REFERENCES

125

5.0 SUMMARY

135

My sincere and heartfelt gratitude is extended to the following persons and institutions without whose support this study could not have been accomplished.

• Prof. W. J. Swart for his excellent supervision, guidance and encouragement during the course of the study period.

• Dr. S. Koch for her constant guidance, help and encouragement during the study period.

• Dr. E. G. Groenewald, Department of Botany and Genetics, for his indispensable assistance in the execution of laboratory techniques pertaining to Chapter 3. • Mrs Marie F. Smith from ARC-Biometry Unit for helping me in the analysis of

data.

• Personnel the Department

of

Plant Pathology for providing constant support during executing of the project, particularly Prof. Z. A Pretorius, Ms C. M. Bender and Ms W. M. Kriel.

• My colleague, Mr. Michael Tecle for his helpful assistance throughout my research.

ACKNOWLEDGMENTS

• The Government of Eritrea, Ministry of Agriculture-Department of Research and Human Resources Development and University of Asmara-Department of Human Resources Development are acknowledged for allowing me to pursue this study and for financial sponsorship.

e My parents, sister, brothers, relatives and friends for the love, encouragement

Table 2.1 Table 2.2 Table 2.3 Table 2.4 Table 3.1 Table 3.2 Table 3.3 ILIST OF TABLES

Host range, localities and dates from which the seven

C. gloeosporiodes isolates where collected. 90

Disease severity expressed by seven C. gloeosporioides isolates on

16 Lupinus cultivars 91

Analysis of variance of anthracnose severity recorded in Trial 1 92 Analysis of variance of anthracnose severity recorded in Trial 2 93 Pure a-amylase assay for the determination of the conversion

factor (CF) from starch samples at incubation temperatures of 20

and 25°C for 5 min and 3 min respectively ' ll1

Average change in optical density (OD) following the effect of GA3-concentration on a-amylase released by barley seed

incubated at 20°C and 25 °c for 5 min and 3 min respectively 112 Total dry mass of mycelium (g), average ug a-amylase

produced,and the extrapolated average GA3-equivalents (jig/g)

Figure 2.1 Figure 3.1 Figure 4.1 Figure 4.2 Figure 4.3 Figure 4.4 Figure 4.5 lUST OF FIGURES

Disease severity assessment diagram used in this study 91 Relationship between gibberellic acid (GA3) concentration

and a-amylase release from barley half-seeds (standard curve)

.113

Germinating conidia of C gloeosporioides at 6 hpi on

susceptible (A) (Bar

=

l Oum, 3,000 x) and less susceptible (B)

(Bar

=

l Oum, 2,500 x) lupin leaves: Spore (s);

germtube (gt); and septurn (sp) 129

Fully formed appressoria of

C

gloeosporioides at 24 hpi on

susceptible (A) (Bar

=

l Oum, 2,500 x) and less susceptible (B)

(Bar

=

l Oum, 2,500 x) lupin leaves: Appressoria (a);

germtube (gt); septurn (sp); and spore (s)

.1

30

Appressoria of C gloeosporioides starting to collapse at 72 hpi on susceptible (A) (Bar

=

l Oum, 2,200 x) and less susceptible (B) (Bar

=

l Oum, 1,700 x) lupin leaves: Collapsing appressoria (ea);

germtube (gt); and spore (s) 131

Development of infection peg at 48 hpi (200

xL

132

Collapsed appressorium of

C

gloeosporioides at 96 hpi on

susceptible (A) (Bar

=

l Ourn, 750 x) and less susceptible (B) (Bar

=

101lm, 2,200 x) lupin leaves: Collapsing appressorium (ea);

Figure 4.6 Development of an infection vesicle and primary hypha on

leaves of lupin at 96 hpi on susceptible (A) (Bar

=

101lm, 2,000 x) and less susceptible (B) (Bar

=

l Oum, 1,800 x) lupin leaves:

Sub-epidermal infection vesicle (iv); primary hypha (ph)

GENERAL

INTRODUCTION

Colletotrichun gloeosporioides (Penz.) Penz. & Sacc. causes anthracnose on a wide range of hosts including, legumes and fruits, that are of particular economic importance (Manners et al., 2000). Anthracnose of lupin (Lupinus spp.) is a destructive disease although consensus on the precise identity of the pathogen has not been reached (Gondran, 1994; Koch, 1996; Reed et al., 1996; Yang & Sweetingham, 1998; Lardner et

al., 1999; Johnston, 2000). Resistance to anthracnose in cultivars of Lupinus

angustifolius has been identified (Cowling et al., 1999). Unfortunately L. albus, which is

desirable for its higher yield and protein content, is very susceptible to the disease (Gondran & Pacault, 1997; Cowling et al. 1999)

The epidemiology of the disease (Gondran & Pacault, 1997) as well as variability in pathogen virulence and host-plant resistance, have been widely studied (Cowling et

al.. 1999). Knowledge of the race composition of pathogens and their geographic distribution may help not only in the development of effective breeding programs, but also in averting the potential risk of crop damage. Monitoring isolates for possible shifts in pathogenicity may also help in the design of control strategies for the disease. An understanding of the mechanisms of penetration and infection of the pathogen will also be very useful in the development of sustainable disease control strategies (Bailey et al., 1992). This applies particularly to pre- and post-penetration, the host-pathogen interaction and biochemical and physiological aspects of pathogenesis.

This thesis represents a compilation of four independent manuscripts based on research conducted over a period of two years. Each chapter is a separate entity within the framework of the lupin-Colletotrichum gloeosporioides interaction and intended to address certain relevant problems. The first chapter is a review of literature on the infection process of Colletotrichum species. The South African lupin industry relies on

cultivars of L. angustifolius. Any shift in the pathogenicity of C. gloeoe.sporioides is likely to be devastating to cultivation of the crop in South Arica. It is in this context that the second chapter sets out to assay the virulence of seven field isolates of C.

gloeosporioides collected from the Western Cape Province, South Africa against 16 lupin cultivars.

Chapter 3 deals with the production of a gibberellin-like substance present in culture filtrate of C. gloeosporioides isolates collected in South Africa. The possible effect this may have on symptom expression is discussed. The fourth chapter of this thesis entails a preliminary examination of the infection process of C. gloeosporioides in lupin. Comparisons between histological reactions to the pathogen of susceptible (Kmutant) and less susceptible (Wonga) cultivars using fluorescence-light microscopy and scanning electron microscopy (SEM) are made. It is hoped that my study will contribute to a better understanding and knowledge of the infection process of lupin anthracnose and its control.

REFERENCES

Bailey, J. A., Q'Connell, R. J. Pring, R. S. & Nash, C. 1992. Infection strategies of

Colletotrichum species In: Bailey, J. A. & Jeger, J. M (eds.), Colletotrichum:

Biology, Pathology, and Control, CAB International, Waillingford, pp., 88-120.

Cowling, W. A., Buirchell, B. J., Huyghe, C., Koch, S. H., Neves Martins, 1. M., Romer, P., Sweetingham, M. W., Van Santen, E., Von Baer, E., Wink, M. & Yang, H.

1999. International evaluation of resistance to anthracnose in lupins. International Lupin Congress, Germany, July 1999.

Gondran, J. 1994. Anthracnose. In: Gondran, J., Bournovill, R. & Duthion, C. (eds.),

Identification of Diseases, Pests and Physical Constraints in White Lupin.

INRA

Editions, France, pp., 20-21.

Gondran, J. & Pacault, D. 1997. White lupin anthracnose. Phytoma 49: 28-31.

Johnston, P. R. 2000. The importance of phylogeny in understanding host relationships within Colletotrichum. In: Prusky, D., Freeman, S. & Dickrnan, M. B. (eds.),

Colletotrichum. Host Specificity, Pathology. and Host-Pathogen Interaction,

ASP Press, St. Paul, Minnesota, USA, pp., 21-28.

Koch, S. H. 1996. Colletotrichum species on dry beans and lupins in South Africa. Ph.D thesis, University of Pretoria, South Africa.

Lardner, R., Johnston, P. R., Plummer, K. M. and Pearson, M. N. 1999. Morphological and molecular analysis of Cofletotrichum acutatum sensu lato. Mycological

Research 103: 275-285.

Manners, 1. M., Stephenson, S-A., He, C. & Maclean, D. J. 2000. Gene transfer and expression in

C.

gloeosporioides causing anthracnose on Stylosanthes. In: Prusky, D., Freeman, S. & Dickrnan, M. B. (eds.), Colletotrichum. Host

Specificity, Pathology, and Host-Pathogen Interaction, ASP Press, St. Paul. Minnesota, USA, pp., 180-194.

Reed, P. J., Dickens, J. S. W. & O'Neill, T. M. 1996. Occurrence of anthracnose (C

acutatum) on omamentallupin in the United Kingdom. Plant Pathology 45: 245-248.

Skipp, R. A., Beever, R. E., Sharrock, K.R., Rikkerink, E. H. A. & Templeton, M. D.

1995. Colletotrichm. In: Kohrnoto, K., Singh, U. S. & Singh, R. P. (eds.),

Pathogenesis and Host Specificity in Plant Diseases: Histopathological Biochemical, Genetic and Molecular Bases, Elsevier Science Ltd. ·UK, pp.,

119-143.

Yang, H. A. & Sweetingham, M. W. 1998. The Taxonomy of Colletotrichum isolates associated with lupin anthracnose. Australian Journal of Agricultural Research. 49: 1213-23.

CHAPTER 1:

A REVIEW OF THE INFECTION AND PENETRATION PROCESS IN THE

1.1 INTRODUCTION

Colletotrichum Corda (anamorph) is a large fungal genus and one of the most important genera of plant pathogenic fungi in the sub-tropical and tropical regions of the world (Bailey et al., 1992; Sutton, 1992). lts teleomorph is the ascomycete genus,

Glomerella von Schrenk and Spaulding. Colletotrichum spp. cause pre- and post-harvest diseases on a wide range of plant species (Jeffries & Dodd, 1990) and have a significant impact on agriculture worldwide through their capacity to cause economic crop losses. Symptoms that are attributed to Colletotrichum spp. are commonly known as anthracnose and typically include depressed black lesions, which are subcuticular or angular and from which erumpent pink spore masses develop (Sutton, 1992).

An understanding of the processes that determine successful pathogenesis is a prerequisite for conducting biochemical and molecular studies relevant to plant pathology. The information can serve as a useful basis for fundamental research in physiological, genetic and molecular aspects of plant pathology that can be exploited by developing novel strategies for disease control (Skipp et al., 1995). Knowledge of the entire infection process can provide information on which to base forecasting models and aid in the development of appropriate agricultural practices. It can also elucidate certain facets of the pathogen's life cycle that could be exploited in optimizing disease control strategies (Bailey et al., 1992).

The present review deals specifically with the morphogenesis of infection structures and the penetration process pertinent to Colletotrichum species and presents recent findings related to these topics (Muirhead, 1981; Bailey et al., 1992; Skipp et al.,

1995). Although an attempt will be made to review biotic factors that affect spore germination and appressorium formation, environmental factors that affect these processes, is beyond the scope of this review.

1980).

1.2 THE INFECTION PROCESS

1.2.1 Spore survival and dissemination

Two major sources of inoculum for Colletotrichum are conidia produced in acervuli and ascospores produced in perithecia (Bailey et al., 1992). Sclerotia are also formed by Colletotrichum truncatum (Schw.) Andrus &Moore in culture and on soybean plants (Khan & Sinclair, 1992) and by Colletotrichum coccodes (WaIler.) Hughes on

tomato (Hornby, 1968). Sclerotia resemble old acervuli and their production allows the fungus to over winter in soil and crop debris (Khan & Sinclair, 1992). Dissemination of spores from young acervuli occurs via water droplets, whilst wind can distribute dry spore masses arising from older acervuli and ascospores (Nicholson & Moraes, 1980). Conidia and ascospores are encased in a moist hydrophilic mucilaginous matrix that maintains conidium viability by acting as a powerful anti-desiccant to ensure spore survival (Nicholson & Moraes, 1980; Ramadoss et al., 1985; Louis et al., 1988; Van Dyke & Mims, 1991). A high molecular-weight glycoprotein, which protects spores from desiccation, is the primary component responsible for the anti-desiccant properties of the mucilaginous matrix of Colletotrichum graminicola (Ces.) Wilson (Bergstrom & Nicholson; 1981; Ramadoss et al., 1985). The germination rate of washed, matrix free conidia is negligible, whilst unwashed conidia germinate at the same rate as freshly harvested conidia (Louis et al., 1988). Removal of the mucilaginous, water-soluble spore matrix from spores prior to storage at any relative humidity for 24 hrs results in a significant reduction in the viability of

C.

graminicola spores (Nicholson & Moraes,

Spores of many Colletotrichum species germinate readily after dispersal, but very poorly, or not at all, at high concentrations. This phenomenon, known as self-inhibition or auto-self-inhibition, occurs commonly enough among fungi to be considered a

general rule. It is thought to be an ecological adaptation that ensures spatial and temporal distribution of the species (Gottleib, 1973). The spore matrix also contains sufficient levels of inhibitors to prevent germination (Loius et al., 1988; McRae & Stevens, 1990; Mondal & Parbery, 1992). An endogenous self-inhibitor, responsible for reduced germination under high concentration, was reported from conidia of

Colletotrichum gloeosporiodes (Penz.) Sacc.

f

sp. jussiaea, (Meyer et al., 1983; Lax et al., 1985). Chemical investigation of the self-inhibitor revealed dihydro-5-hydroxy-5 [(8 pentyl-2-oxocanyl)-acetyl)-2(3H)-furanone, for which the trivial name 'gloeosporone' was proposed (Meyer et al., 1983). However, Mondal & Parbery (1992) indicated that 'gloeosporone' is mostly present in the spore matrix. Recently Tsurushima et al. (1995)

re-examined the same isolate according to the procedure reported by Meyer et al.

(1983), and could not detect 'gloeosporone'. The authors instead detected three active principles and identified these self-inhibitors as, (E) and (Z)-3-ethylidene-l,

3-dihydroindol-2-one and (2R)-(3-indolyl) propionic acid as CG-SI (C

gloeosporioides-inhibitor) 1, 2 and 3. The inhibitor, CG-SI 1 and 2 inhibited initial germtube emergence from conidia (Tsurushima et al., 1995). Another inhibitor, 'mycosporine-alanine' prevents germination of C graminicola within acervuli and ensures that ungerminated, viable spores can be dispersed across the leaf surface so that secondary infection can occur (Leite & Nicholson, 1992).

The importance of the matrix as a source of nutrients, energy and enzymes for germinating spores has often been demonstrated (Bergstrom & Nicholson, 1981; McRae & Stevens, 1990; Mondal & Parbery, 1992). Tu (1983) reported that the rapid decrease in viability of spores of

C

lindemuthianum (Sacc. & Magn.) Br. & Cav. under wet conditions might be attributed in part to the loss of the mucilaginous water-soluble matrix of conidia. Studies of the composition of the matrix revealed the presence of

glycoprotein with carbohydrates and amino acids (Ramadoss et aI., 1985; Nicholson el

al., 1986) having a high affinity for phenols. This may represent one mechanism

through which fungi inactivate toxic phenolic compounds in their environment (Nicholson ef al., 1986). Proline-rich proteins in the matrix of

C.

graminicola protect

conidia from host-derived toxic phenolic compounds, including p-coumaric acid and ferulic acid that accumulate in water (Nicholson et al., 1986, 1989). Water leachates

from lesions were inhibitory to

C.

graminicola spores only in the absence of mucilage

(Nicholson et al., 1986). O'Connell et al. (1992) reported N-acetyl glucosamine

(GlcNAc) and a-linked mannose or glucose residues depending on the fungal species. The materials appeared to be secreted by conidia during germination, since conidia were washed twice by centrifugation and suspension in water prior to applying them to slides.

The conidial matrix of certain Colletotrichum species plays an important role in

the infection process itself because it contains enzymes that are known to be involved in the penetration of plant surfaces. Pectin esterase and cellulase (McRae & Stevens, 1990) have been reported in the conidial matrix of Colletotrichum orbiculare (Berk & Mont.) Arx. Invertase, cellulase and pectinase were detected in the conidial matrix of

C.

gloeosporioides (Louis

&

Cooke, 1985), while the conidial matrix of

C.

graminicola

contained invertase (Bergstrom & Nicolson, 1981), ,B-glucosidase (Ramadoss et al.,

1985), cellulase and a non-specific esterase (Nicholson & Moraes, 1980; Ramadoss et aI., 1985). It is possible that these enzymes may cleave the phenolic esters and

glycosides freeing the less water-soluble aglycones and making them more available for binding to the extracellular proline-rich proteins of the mucilage (Nicholson et aI.,

1986). The production of larger lesions by older unwashed spores compared to washed spores, suggests that the matrix does indeed play a role (Mondal & Parbery, 1992). The addition of conidial matrix to inoculum of

C.

orbicular significantly increased

anthracnose development in Xanthium spinosum (McRae & Stevens, 1990) while removal of the matrix from inoculum of

C.

gloeosorioides reduced disease severity, the

effect being attributed to a reduction in germination (Louis & Cooke, 1985).

It has been suggested that any inhibitory effect the spore matrix might have will be more pronounced at high spore concentrations, and more beneficial to mass spore survival than that of individual spores (McRae & Stevens, 1990). In spite of confirmation that the spore matrix can influence spore behavior, there is little knowledge of the mechanism involved or its ecological significance (Mondal & Parbery, 1992). The physiological basis for the role of the spore matrix was said to be at cellular and not sub-cellular level (Louis

&

Cooke, 1985). Although nothing is known of the persistence of the spore matrix following spore dispersal, it has been suggested that it may be attenuated to a stage where its effects are negligible (Louis & Cooke, 1985). In the absence of rain or impacting water droplets, mucilaginous spore masses become dry and are then wind dispersed (Louis et al., 1988).

1.2.2 Spore germination and appressorial formation

1.2.2.1 Spore germination

Ungerminated conidia are non-septate and single celled (Skoropad, 1967; Manadhar et al., 1985; Smith et al., 1999; Latunde-Dada et al., 1996; Latunde-Dada et

al.,

1999) but swell (Politis & Wheeler, 1973) and their nuclei divide mitotically prior to germination (Skoropad, 1967; Parbery, 1981;Van Dyke & Mims, 1991). Most germinating conidia have one septum (Jeffries & Dodd, 1990; Latunde-Dada et al.,

1996) but in some rare cases up to three septa have been observed (Skoropad, 1967; Latunde-Dada et al., 1999; Smith et al., 1999). Generally, Colletotrichum species begin to germinate 2-6 h after imbibition (Milholland, 1982; Van Dyke & Mims, 1991;

Q'Connell et al., 1993; Byme et al., 1997; Wei et al., 1997; Pring et

al.,

1995; Smith

er

al., 1999). However, in

C.

lindemuthianum, germination was reported 18 h after imbibition on Phaseolus vulgaris (O'Connell et al., 1985) and 12 h on French bean

(Mercer et al., 1975).

The number of germtubes produced and their position of emergence differ from species to species. Smith et a/. (1999) reported that germtubes developed randomly

from points on the conidia in Colletotrichum dematium (Fr.) Grove. The emergence of a single, lateral germtube near the end of a conidium was reported in Colletotrichum

truncatum (Van Dyke & Mims, 1991), C. gloeosporiodes (Morin et al., 1996), and

C.

dematium (Smith et al., 1999). .Skoropad (1967) reported that most spores of C.

graminicola produce two short (1-2Ilm) germtubes from each end of the conidium although two or three germtubes from each cell have also been observed, especially when germination occurs on nutrient media.

1.2.2.2 Spore adhesion

Adhesion of an infecting spore is an essential component for fungal infection to occur and may be considered as a factor determining the virulence of many pathogens (Nicholson & Epstein, 1991). Adhesion ensures that a pathogen remains in contact with its host for as long as it is necessary for penetration to occur. It also firmly attaches the infection hypha to a site where penetration, whether mechanical or enzymatic, can be achieved (Bailey et al., 1992).

The adhesion of Colletotrichum conidia occurs during a number of morphogenic stages. Conidia of Colletotrichum species can adhere to the host surface prior to germination (Young & Kauss, 1984; Mereure et al. 1994a) or after germination (Mercure et al., 1994a, b; Sela-Buurlage et al., 1991). Rapid adhesion increases the chances of disease development. In nature, conidia of Colletotrichun species are

dispersed in water droplets that easily run off leaf surfaces (Sela-Buurlage et al.. 1991). Louis et al. (1988) suggested that in the absence of rain or impacting water droplets, mucilaginous spore masses become dry and are then wind dispersed. Thus, early adhesion of ungerminated or· germinated conidia ensures that propagules are not removed from the leaf (Mercure et al., 1994b) or any other hydrophobic surfaces (Sela-Buurlage et al., 1991) either by additional rainfall or by wind.

Considerable speculation has surrounded the role of the spore matrix in adhesion. In C. graminicola, removal of the mucilage does not prevent adhesion of ungerminated conidia to the host surface suggesting that it is superfluous (Mercure et al, 1994a, b). Evidence suggests that the adhesive material is released on contacting the surface of the host. Colletotrichum species display better adhesion on hydrophobic than hydrophilic surfaces (Young & Kauss, 1984; Sela-Buurlage et al., 1991; Mereure et al., 1994b). In C. graminicola (Mercure et al., 1995) and C. gloeosporioides (Jones et al., 1995), the release of protein exudates appears to consolidate the initial hydrophobic attachment of conidia. Ungerminated conidia of Colletotrichum musae (Berk. & Curt.)' Arx. (Sela-Buurlage et al., 1991; Mereure et al.. 1994a) and C. graminicola (Sugui et al.. 1998) apparently produced a proteinaceous adhesin. Proteinaceous adhesin from C. musae was produced more than once prior to germtube emergence, since conidia that were protease-E-treated and then washed, regained adhesiveness. Conidia that were maintained in suspension by constant mixing remained capable of adhesion to the substratum for several hours prior to germtube emergence. This may be due to new adhesive material being produced during this period (Sela-Buurlage et al., 1991). Mereure et al. (1994a) reported that when conidia and germtubes were lifted from the host surface, a portion of the leaf cuticle in contact with the fungus was also removed.

Opportunities for detecting a broad range of molecules in the infection structures of Colletotrichum species have been provided by the use of lectin labeling and monoclonal antibodies (MAbs). By using MAbs, glycoproteins with some adhesive property have been found in extracellular matrix (ECM) produced by C

lindemuthianum. A set of glycoproteins that are very abundant in the fibrillar sheath surrounding germtubes were recognized by MAb UB22 (Pain et aI., 1992; O'Connell et

al., 1996) while UB26 recognized two high molecular weight glycoprotiens present on

the surface of germtube and appressoria (O'Connell et aI., 1996: Pain et aI., 1996).

MAb UB31 was also identified on ECM components specific to germ tubes and appressoria (O'Connell et al., 1996). The glycoprotein recognized by UB26 appears to be firmly attached to the fungal wall (O'Connell et aI., 1996). It adhered to the glass substratum on which the germtube grew to the extent that it could not be dislodged by ultrasonication (Pain et al., 1996). UB27 also recognized 48-kDa glycoprotein present in the upper domed regions of plasma membrane of appressoria in C lindemuthianum and was designated as CLA 1 (C lindemuthianum Appressorium 1) (O'Connell et al.,

2000).

In C lindemuthianum MAb, UB20, recognized a glycoprotein on the outer surface of the spore-coat and to a lesser extent at the plasma/cell wall interface (Hughes

et al., 1999; O'Connell et al., 1996). Western blotting with UB20 showed 110 kDa as a

major component of the glycoprotein. At a low concentration, UB20 inhibited attachment in an antigen-specific manner. Polystyrene microspheres bound selectively to the Il0kDa glycoprotein in western blots, providing further evidence that this component could mediate interactions with hydrophilic substrata (Hughes et al., 1999).

O'Connell et al. (2000) reported that a plasmolysed appressorial cell remained in close

glycoprotein may act as an integral molecule, binding the plasma memebrane to the cell wall. However, it remains unclear whether the retracted portions of membrane correspond precisely with the domain lacking CiA 1.

Characterizing changes on the surface of Colletotrichum species during differentiation of infection structures in vitro and infection of host tissue using lectin cytochemistry is widely used (O'Connell et al., 1992). Recently, lectin analysis showed a high molecular weight glycoprotein attached to a hydrophobic substratum using ECM released from conidia and germlings of C. graminicola (Sugui et al., 1998). Earlier Marks et al. (1965) reported that a hyaline mucilaginous substance apparently held the lower appressorial wall firmly to the host cuticle in C. gloeosporiodes. Mature appressoria, for example, were not easily dislodged from the leaf surface of Populus

tremuloides when washed with a gentle stream of water. Sugui et al. (1998) reported that the removal of appressoria of C. graminicola revealed an unstained gold/silver region surrounded by a zone that stained intensely for protein. The area that did not stain positively for protein appears to be the contact interface of the appressorium with the substratum. Examination at high magnifications revealed that the unstained area represents the site of the penetration pore. The association of carbohydrates with the secreted adhesin was demonstrated with different lectins. Mereure et al., (1995) reported that the material released by ungerminated conidia of C. graminicola at the conidium-substratum contact interface contains glycoproteins. However, the clear zone that immediately surrounds conidia is composed of carbohydrates rather than protein as it was labelled by fluorescein isothiocyanate-concanavalin aggulutinin (FITC-ConA) and fluorescein isothiocyanate-ienis cu/inaris aggulutinin (FITC-LCA)· but not by gold/silver stain. Furthermore, since both these lectins bind to glucose and/ or to mannose, the results confirm that the material contains these two sugars.

Adhesive competence depends both on conidia and the leaf surface. There are few reports of the influence of conidium and leaf age on adhesion. Mereure ef al.

(1994a) reported a significant difference in the ability of 14-day-old conidia to adhere to the leaves of 5- and 8-week-old plants. Adhesion approached 30% of the conidial population by 30 min on 5-week-old plants. In contrast, adhesion of conidia on 8-week-old plants reached a maximum of only 20% by 2 h. The adhesive competence of ungerminated conidia decreased as conidia aged (Leite & Nicholson, 1992; Mereure el

al., 1994b).

Reports on the role of spore metabolism in adhesion in different species of

Colletotrichum are conflicting. Sela-Buurlage ef al. (1991) claimed adhesion is an

active metabolic process, as conidia killed with UV light, formaldehyde or heat were significantly less adhesive than non-treated controls. Spore metabolism may also be required for conidia to remain adherent since adhered conidia became detached after exposure to a lethal dose of UV light (Sela-Buurlage ef al., 1991). Young & Kauss (1984) reported that respiratory inhibitors significantly reduced adhesion of C.

lindemuthianum spores while Mereure ef al. (1994b) reported that respiratory inhibitors

had no apparent effect on the adhesion of C. graminicola conidia. Exposure of conidia

to either low (4°C) or high (50°C) temperatures also did not significantly affect adhesion after 30 minutes. These findings suggest that active respiration may not be required for adhesion of ungerminated conidia.

Whether the adhesive material is released before or after conidia contact the substratum is open to question. In C. graminicola, Mereure et al. (1994a) proposed that

adhesive was present pre- and post-adhesion. Autoclaving conidia prevented further production of adhesive but did not destroy the adhesive already present on the surface.

Furthermore, treatment of conidia with either a protein synthesis inhibitor or a glycoprotein synthesis and transport inhibitor significantly reduced adhesion.

1.2.2.3 Appressorial formation

All species of Colletotrichum produce simple appressona although certain germinating spores do not produce appressoria (Porto et al., 1988). Appressorium initiation may occur soon after germtube emergence, with some species producing a sessile appressorium (Morin et al., 1996; Smith et al., 1999). Germtubes sometimes also branch profusely to produce multiple appressoria (Latunde-Dada et al., 1999) while

subtended appressoria commence as a terminal swelling of the germtube and soon separate from the spore by means of a sept urn (Skoropad, 1967; Politis & Wheeler, 1973; Mercer et al., 1975; Morin et al., 1996).

Walls of appressoria are composed of an inner electron opaque layer and an outer electron lucent layer (Xuei et al., 1988) and are usually a two- (Mercer et al., 1975; Bell & Wheeler, 1986; Coates et al., 1993) or three-layered structure (Mould et ai, 1991a). Appressoria initially are hyaline but become darker when melanisation occurs (Muirhead & Deverall, 1981; Coates et al., 1993; Morin et al., 1996; Byrne et al., 1997). Melanin provides a typical dark appearance and may protect them from irradiation, but the pigment may also play a crucial role in the penetration process (Katoh et al., 1988; Howard & Ferrari, 1989). Melanin strengthens the appressorial wall to support the high internal hydrostatic pressure necessary for penetration of the plant cuticle and also determines the direction in which the infection peg emerges (Kozar & Neolitzky, 1978; Kubo & Furusawa, 1985). Dormant appressoria can 'withstand desiccation and high temperatures and ensures active 'growth when conditions are more favorable. The appressoria of C. gloeosporioides can survive at least 6 months on glass cover slips; but

when plated on potato-dextrose agar (PDA), they resume normal growth and eventually sporulate (Holrnstrom-Ruddick & Mortensen, 1995).

Most species of Colletotrichum appear to form appressoria only when germtubes touch rigid surfaces. In C. truncatum germ tube length is variable and appears to be related directly to moisture conditions, with wetter conditions favoring longer germtubes and on nutrient agar germtubes usually attain a considerable length before appressoria are formed, whereas on leaves, appressoria are formed immediately or after very limited elongation (Skoropad, 1967; Van Dyke & Mims, 1991). Dickman (2000) has reported

Colletotrichum trifolii Bain & Essay. ras protein (CT-Ras) that regulates a signal transduction pathway that senses and responds to nutrients. Under nutrient-limiting situations a wild-type cells arrest vegetative growth (hyphal elongation) and differentiate (conidate), whereas cells expressing constitutively active CT-Ras continue vegetative growth, do not differentiate and, importantly, are impaired in polarized, systemic growth.

In C. graminicola (Po litis & Wheeler, 1973) and in C. trifolii (Mould et al., 1991a), each appressorium appears to contain a single nucleus with a prominent nucleolus packed with large number of free ribosomes, lipid bodies, mitochondria and glycogen before the appearance of the penetration peg. Mercer et al. (1975) reported only ribosomes and small lipid droplets in appressoria of C. lindemuthianum. In contrast, Skoropad (1967) reported the appearance of two nuclei in appressoria of C.

graminicola, prior to their movement into penetration hyphae.

The lower portion of the appressorial wall contains a central germ pore, which provides an opening for the emerging infection peg/hyphae (Brown, 1977; Politis & Wheeler, 1973; Coates et aI., 1993). Dissolution of the appressorial wall to form the pore was reported in C. graminicola (Polities & Wheeler, 1973). The infection peg is

very thin when it passes through the cuticle, but it increases in diameter once inside the host cell (O'Connell et al., 1985). It penetrates the host cell after emerging through the appressorial pore something, which has not been observed in all species (Po litis & Wheeler, 1973).

An appressorial cone of varying shape normally surrounds the germpore and the wall of the infection peg appears to be an extension of the cone (O'Connell & Ride, 1990; Mould et al., 1991a). It may act to focus hydrostatic pressure on the site of penetration (Wolkowet ai, 1983) and could therefore, be considered as part of the

infection hypha (Bailey et al., 1992). Synthesis of the appressorial cone probably represents one of the initial stages in formation of infection hypha (Brown, 1977).

Ultrastructural studies have illustrated a fibrillar layer around the germtubes and mycelium of Colletotrichum atramentarium (Berk. & Br.) Taubenh (Griffiths & Campbell, 1973), C. graminicola (Kozar & Netolitzky, 1978) and C. truncatum (Van Dyke & Mims, 1991) as well as the germtubes and appressoria of C. lindemuthianum (O'Connell et al., 1996). The walls of ungerminated and germinated conidia of C.

truncatum (Van Dyke & Mims, 1991) are coated with fibrillar material. The spore coat in some cases appears to consist of preformed structures present in ungerminated, unimbibed conidia (Van Dyke & Mims, 1991). The maturation of appressoria was also associated with a thin, intensely fluorescent secretion of mucilage that was clearly distinct from, and external to, the melanized appressorium. The pattern of lectin binding to mucilage from different Colletotrichum species suggested the presence of polysaccharides or glycoproteins containing /]-1, 4-linked GlcNAc, N-acetyl galactosamine (GaINAc) or galactose and a-linked mannose or glucose residues (O'Connell et al., 1992).

In

C.

trifolii on Medicago sativa L. (Mould et al., 1991a; Mercer et al., 1975) and in

C.

lindemuthianum on cowpea, Vigna unguiculata (Bailey et al., 1990),

appressorial maturation culminated in the migration of the conidial cytoplasm from the conidium to the appressorium via the germtube. When devoid of cytoplasm. the spore and germtube become vacuolated and conidia distal to the germtube appear as a non-staining 'ghost-cell' empty of identifiable cytoplasm, which subsequently collapses, Prior to penetration, vacuoles when present are small and scattered (politis & Wheeler, 1973). In some Colletotrichum species, host penetration and colonization occurs immediately after appressorium formation, In other species, environmental conditions may be fungi static prior (Binyamini & Schiffmann-Nadel, 1972; Parbery & Emmett, 1977) or post (Verhoeff, 1974; Chau & Alvarez, 1983; Coates et al., 1993; Prusky et al., 1998) penetration of the host. This aspect will be dealt with in para. 3,3.

1.2.2.4 Germination and appressorium formation

Emmett & Parbery (1975) suggested that in most species appressoria develop as a matter of course or as an end point to germination, provided the external environment at the plant surface is conducive to appressorium formation, Several factors interacting at the plant surface may inhibit or stimulate appressorium formation, A number of physical and chemical signals necessary for induction of appressoria formation have been reported, Various chemical substances associated with the plant surface are likely to exert an effect on the formation of appressoria. Chloroform extracts from sugar beet leaves on

C.

acutatum (Parbery & Blakeman, 1978) and anthranilic acid from banana on

C.

musae (Swinburne, 1976) stimulated appressorium formation, Anthranilic acid is converted to 2,3-dihydoxybenzoic acid that was usually the most effective stimulant. The possibility is that anthranilic acid is not a germination stimulant per se but active only because of its conversion to 2,3-dihydoxybenzoic acid (Harper & Swinburne, 1979)

Chemical stimuli present in epicuticular wax are implicated in conidial germination and appressorium formation. Conidia of C gloeoesporioides are induced to germinate and differentiate to form appressoria by chemicals centered in the wax on the host surface (Prusky et aI., 1991; Podila et aI., 1993). Chemical signals are also emitted during fruit ripening by the hormone ethylene (Flaishman et aI., 1995). Solid contact surface is the principal requirement and stimulator for the morphogenesis of conidia from C gloeosporioides (Kim et al., 1998; Liu & Kolattukudy, 1998) and C trifolii (Buhr & Dickman, 1997) in response to surface waxes and ethylene. Conidia resting on either a hard hydrophilic surface (glass) or a hard hydrophobic surface responded to the chemical signals only between 2 and 4 h after the initiation of contact with the hard surface (Flaishman et al., 1995; Hwang & Kolattukudy, 1995). Using differential-display methods, chip] (Colletotrichum hard-surface induced protein 1 gene), which encodes an ubiquitin-conjugated enzyme, was expressed during the early stages (2 h) of hard surface treatment (Liu & Kolattukudy, 1998).

Cell signaling pathways operating in spore germination and appressorium formation of Colletotrichum species have been studied by monitoring the expression of genes encoding putative signaling components, or their inhibition. In C trifolii, appressorium formation, in contrast to conidial germination and germtube formation, strongly depends on Ca+2. The disturbance of calcium homeostasis, by ethylene-bis (oxyethylenenitrolo) tetra-acetic acid (EGTA) or calcium channel blockers, had a negligible effect on conidial germination and germtube growth, but markedly impaired appressorium development (Dickman et al., 1995; Warwar & Dickman, 1996). Moreover, calmodulin (CaM) inhibitors affect both germination and differentiation implying that the Ca+2/ calmodulin (CaM) signal transduction pathway is important in the early development of C trifolii on the plant host surface (Dickman et al., 1995;

Warwar & Dickman, 1996). Rather than a transient increase in Ca+2, calcium cycling across the membrane is required for differentiation (Dickman et al., 1995; Warwar &

Dickman, 1996). RNA analysis of signal-transducing genes from C. trifolii, including

genes for a serine-threonine kinase (TB3), calmodulin and protein kinase C, showed maximum transcription of all the three genes in conidia prior to or during germ tube morphogenesis. TB3 and calmodulin gene transcription peaked during germtube morphogenesis (Buhr & Dickman, 1997). A putative

CaM

kinase (CaMK) cDNA of C.

gloeosporioides was cloned with transcripts from hard surface treated conidia. The

inhibition of this enzyme by KN93 (20 microM) inhibited germination and appressorium formation, blocked melanization and caused the formation of abnormal appressoria (Kim

efal., 1998). Yang & Dickman (1997) suggested the involvement of cAMP and

cAMP-dependent protein kinase pathways. Both conidial germination and appressorial differentiation were impaired using specific inhibitors of cAMP-dependent protein. Using a pharmacological approach, including 8-Br-cAMP, sodium fluoride and 3-isobutyl-1-methylexanthine, all of which increase endogenous cAMP levels, showed the induction of appressorial differentiation on a non-inductive surface (1.5% water agar) (Yang & Dickman, 1997). In C. trifolii on alfalfa, a single copy gene (Ct-PKAC) encoding the catalytic cAMP-dependent protein kinase was isolated (Yang efal., 1999).

Transformants obtained through insertional activation of Ct-PKAC by gene replacement showed a small reduction in the growth relative to the wild type and conidiation patterns were altered. Importantly, PKA-deficient mutants could form appressoria, colonise wounded leaves and produce acervuli suggesting that loss of pathogenicity is most likely due to the failure in appressorial penetration (Dickrnan, 2000; Yang ef al., 1999). In C.

lagenarium CMKl gene encoding a mitogen-activated protein (MAP) kinase regulates germination triggered by both glass surface contact and nutrient. signals. However,

restoration of germination through the addition of yeast extract in CMKl mutant suggests that presence of a bypass pathway that can induce germination independent of

CMKl pathways (Takano et al., 2000). Although it was speculated that the synthesis or

secretion of some extracellular materials shown to be involved in conidium attachment (Mercure et al., 1995) might be impaired by disruption of CMK1, failure of germination in the CMKl mutant partially due to weak attachment of conidia cannot be excluded (Takano et al., 2000). Kim et al. (2000) reported the cloning of a mitogen-activated protein kinase (MEK), designated as CgMEK1, from

C.

gloeosporioides. It was involved in a polarized cell division, with a preferential increase in F-actin in one of the daughter nuclei in response to hard surface contact, septurn formation, germination and differentiation of germ tube to appressoria. Disruption of this gene blocks hard surface-induced cytokinesis at a stage immediately following nuclear division. Thus, in the

CgMEK1-disrupted mutants, the preferential increase in F-actin associated with one of the daughter nuclei and septurn formation does not occur. This loss of polarity prevented germination and hyphal development and consequently loss of pathogenicity on its natural host. Instead, the CgMEKl mutants exhibited a budding-type of growth leading to the formation of oval cells. Although the exact sequence of signaling events needs to be further elucidated, the aforementioned evidence suggests that signaling pathways are involved during conidial germination and the differentiation of appressoria in Colletotrichum species.

Certain molecular events are triggered in Colletotrichum species by chemical signals from the host. Differential screening of mRNA from non-germinating appressoria that form conidia gave four cDNA clones representing transcripts found only in appressoria forming spores of

C.

gloeosporioides (Hwang et al., 1995; Hwang & Kolattukudy, 1995). Two of these clones' cap20 and cap22 genes were uniquely

expressed during appressonum formation after 4 h exposure of spores to wax. Immunogold labeling with antibodies against cap20 and cap20 proteins showed that both of these gene products are located in appressoria (Hwang et al., 1995; Hwang & Kolattukudy, 1995). Spores of cap20 gene-disrupted mutants germinated and formed normal looking appressoria. Structural changes that might have resulted from the lack of

cap20 protein were not manifested in gross morphological alternations but showed a drastic decrease in virulence on avocado (Hwang et al., 1995). It was concluded that

cap20 protein is necessary to make a functional penetration structure, the mechanism of

which remains to be elucidated.

A number of external factors interacting on the plant surface are likely to influence appressorium formation. Species of Colletotrichum commonly fail to produce appressoria in the presence of abundant exogenous nutrients but germ tubes continue to grow and branch (Emmett & Parbery, 1975; Lenne & Parbery 1976). Mycelial development of

C.

musae, which is much greater on ripe bananas, presumably reflects

higher concentration of nutrients in the leachates (Swinburne, 1976). The deprivation of both amino acids and sugars caused the formation of appressoria in Colletotrichum

acutatum Simmonds (Blakeman & Parbery, 1977). A single factor operating alone, however, is unlikely to be the sole stimulus in vivo. Exposure to either of the two components caused the formation of long germ tubes without appressoria (Blakeman & Parbery, 1977). The requirement for exudates from ripe pepper fruits containing nutrients to stimulate appressorium formation in Colletotrichum piperatum Ell. & Ev. (Grover, 1971) may represent the exceptional adaptation amongst Colletetrichum species to enable infection of ripe fruits to take place. A high percentage germination and appressorium formation was observed on the leaf surface of Stylosanthes guianesis. Appressorium formation per se was neither inhibited nor enhanced and the number of

appressoria produced was directly related to the number of germinated spores (Lenne & Brown, 1991).

Microorganisms that inhabit the phylloplane may also stimulate the formation of appressoria. Species of Bacillus (Lenne & Parbery, 1976) and Pseudomonas (Blakeman & Brodie, 1977; Blakeman & Parbery, 1977), for example, have been shown to reduce germination but promote appressorial formation in C. gloeosporioides and

Colletotrichum dematium (Pers. ex Fr.)

f

sp. spinacea, respectively. Blakeman & Parbery (1977) suggested that Pseudomonas caused this effect by imposing nutrient stress on the fungus, since leaching nutrients from germinating spores could produce a similar effect. Lenne & Parbery (1976) reported a rapid lysis of the germtube and spores of C. gloeosporioides by the bacterium but an inability to lyase the appressorium.

The importance of protein synthesis for spore germination and appressorial formation by anthracnose fungi has also been investigated. Spores of Colletotrichum

lagenarium (Pass.) Ell. & Halst. can germinate 40 min after the start of incubation even if protein synthesis is then inhibited by cycloheximide. Before this time however, protein synthesis is indispensable for spore germination (Furusawa et al., 1977; Suzuki et al., 1981). Morphogenesis of appressoria does not require de novo synthesis of protein after 40 minutes of incubation. When protein synthesis was completely inhibited by cycloheximide after one hour of incubation, appressoria matured in structure but not in function. Appressoria seemed to have no ability to penetrate artificial membranes (Suzuki et al., 1981). In C. trifolii, conidial germination does not require de novo protein synthesis. Cycloheximide had no effect on spore germination but fungal growth ceased following germ tube growth (Dickman et al., 1995). Pascholati et al. (1993) showed that serine esterase inhibitors (Di-ispropyl flourophosphate-DIPF) have little effect on formation of appressoria by C. graminicola either on polystyrene or green leaf

tissue. Appressoria appear to mature even in the presence of DIPF but disease development does not occur. Suzuki et al. (1981, 1982a) concluded that while germling differentiation does not require protein synthesis once germination has started, appressonum maturation does. Cycloheximide, did not inhibit appressonum development, but inhibited the pigmentation of appressoria instead.

1.3 PENETRATION AND POST PENETRATION REACTIONS

1.3.1 Mode of penetration

Colletotrichum species penetrate the plant hosts through natural openmgs, through wounds and/or by direct penetration of the cuticle (Bonnen & Hammerschrnidt,

1989a; Bailey et al., 1992). The most common means of penetration is directly through the cuticle and epidermal cells. It usually occurs after formation of appressoria and the

infection peg (Manadhar et al., 1985; Latunde-Dada et al., 1996; Morin et al., 1996;

Byme et al., 1997; Smith et al., 1999). Direct penetration with an undifferentiated germtube has also been reported in C gloeosporioides (Daquioag & Quimio, 1978; Manadhar et al., 1985; Ogle et al., 1990; Das & Bora, 1998).

Penetration through natural openings by Colletotrichum species has also been reported. Latunde-Dada et al. (1999) reported exclusive entry through stomata using isolate LARS 860 from cowpea preliminarily identified to be Colletotrichum destructivum O'Gara. This species used undifferentiated germtubes. While melanized appressoria were produced abundantly on the host surface, no infection hyphae were observed below them. C dematium on onion (Russo & Pappelis, 1981), C acutatum on guava fruit (Das & Bora, 1998) and C capsici on cotton (Roberts & Snow, 1984) have been reported to penetrate through stomata, although these species are also capable of penetrating the cuticle and cell wall. TeBeest et al. (1978) reported penetration by appressoria of C gloeosporioidesf sp. aeschynomene through the base of the numerous trichornes on the stem on northern joint vetch.

1.3.2 Mechanisms of penetration

Bailey et al. (1992) in their review mentioned the dubiousness of the mechanisms of penetration of the cuticle and epidermal cell walls as did Skipp et al.

(1995). Nevertheless, three mechanisms have been proposed: mechanical and enzymatic, either alone or acting synergistically.

Mechanical force as a mechanism of penetration was demonstrated by the indentation of a host surface associated with appressoria (Xuei et al.. 1988) or the

infection peg (Mercer et al., 1975) of C. lindemuthianum on the host cell wall. Mercer

et al. (1975) reported that apart from a considerable increase in size of the infection

hyphae, there was no visual evidence of cellulose degradation and other components of epidermal cell walls after penetration of the cuticle. Cutin degrading enzymes may thus not be involved (at least not as primary determinants) in penetration and mechanical force may be a more important .factor involved in cuticular penetration by this fungus (Bonnen & Hammerschmidt, 1989a,b). In spite of an apparent inhibition of appressorial formation, conidia of C. lagenarium treated with inhibitors could still cause disease in

etiolated cucumber hypocotyls. The addition of paraoxon, an inhibitor of cutinase, failed to affect the radial growth of C. lagenarium on PDA.

Mechanical penetration is supported by evidence from the research of several groups investigating the importance of melanization of appressoria in mechanical penetration. Research with C. lindemuthianum (Wolkow er al., 1983), C. lagenarium

(Kubo et a!.. 1982; Katoh et a!., 1988) and C. graminocola (Pascholati et al., 1993)

suggests that melanized and structurally sound appressoria may provide the rigidity necessary to support the mechanical force required for penetrating the plant cuticle. Non-melanized appressoria of C. lagenarium, whether due to specific chemical

inhibitors (tricyclazole) or mutation, resulted in a 90-95% reduction in the penetration of nitrocellulose membranes by the fungus (Kubo et al., 1982; Suzuki et al., 1982b; Katoh

Melanization of appressoria provides cell wall rigidity (Kubo el al., 1982; Katol

et al., 1988; Kubo & Furusawa, 1991; Pascholati et al., 1993) and also plays a role in osmosis by allowing a build up of internal hydrostatic pressure (Kubo & Furusawa,

1986; Howard & Ferrari, 1989) necessary for mechanical penetration. Colourless appressoria of the albino mutant of C. lagenarium germinated laterally on host plants or nitrocellulose membranes and consequently could rarely penetrate them. But addition of scytalone, a natural intermediate of melanin biosynthesis, isolated from the color mutant restored both appressorial pigmentation and penetration (Kubo et al., 1982, 1983). Studies to elucidate the melanin biosynthesis pathways using melanin deficient mutants and homologues of known genes have shown that three melanin biosynthesis genes, notably PKS1, SeD1 and THR1, have been cloned and characterized for C. lagenarium

(Takano et al., 1995; Kubo et al., 1996; Perpetua et al., 1996). Polyketide synthase (encoded byPKS1) is involved in the first step (Kubo et al., 1991; Takano et al., 1995).

Subsequent steps consist of dehydration and reduction reactions. The dehydration of scytalone to 1, 3, 8-trihydroxynaphthalene (1, 3, 8-THN) and vermelone to 1, 8-dihydroxynaphthalene are performed by scytalone dehydratase (encoded by SeD 1)

(Kubo et al., 1996). Reduction of 1, 3, THN to vermelone is performed by 1, 3,

8-THN reductase (encoded by THR1) (Perpetua et al., 1996). Melanin is then yielded by

the polymerization and oxidization of 1,8-dihydroxynaphthalene. Takano et al. (2000) have confirmed that transcription of PKS1 and SeDl was tightly linked to conidial

germination process as regulated by CMK 1 gene encoding a mitogen-activated protein (MAP) kinase but not to subsequent appressorium formation. THR1 disappeared when

conidia could not germinate but accumulated the same as the former two when conidia germinated. This was consistent with a previous report that de novo transcripts of the above three melanin biosynthesis genes accumulated 1 to 2 h after the start of conidial

incubation at 24°C, in advance to formation of appressoria and melanization (Takano et

al., 1997).

The role of enzymes during penetration has been examined and two lines of evidence have been used to support the role of cutinase in penetration and infections: a). specific organophosphate inhibitors of cutinase to block infection and b). antibodies raised against cutinase. Application of the potent cutinase inhibitor, di-ispropyl fluorophosphate, along with spores of

C.

gloeosporiodes prevented infection of the host

(Dickman et al., 1982). Pring et al. (1995) reported that the initial penetration of the cuticle of cowpea by Colletotrichum capsid (Syd.) Bud. & Bisby appears to be at least partially enzyme mediated as the penetration pore is well defined with no cuticular debris associated with penetrating hyphae. Similar results were obtained when antibodies to cutinase were included in spore suspensions (Dickman et al., 1982). Neither the antibodies (Dickman et al., 1982; Dickman & Patil, 1986) nor di-ispropyl fluorophosphate suppressed lesion formation when papaya's cutin barrier was breached by a needle prick prior to inoculation. These findings confirm that

C.

gloeosporiodes

can penetrate the cuticular layer of papaya by secreting cutinase (Dickman et al., 1982). At least for certain direct penetrating fungi, enzymes are less important than mechanical force. It is possible that cutin-degrading enzymes play a minor role in reducing the force necessary for penetration by loosening the cutin. This is an activity that may not be essential or detectable (Bonnen & Hammerschmidt, 1989a).

1.3.3 Latency

Latency can be defined as a quiescent or dormant parasitic relationship that a pathogen has with its host and which after a certain period converts to a pathogenic relationship (Muirhead, 1981; Coates et al., 1993). The resumption of pathogenic activity generally happens during major changes in the host's physiological state such as

The transition from a dormant parasitic relationship to an active one usually takes place only when fruits ripen, a phenomenon that is difficult to explain. Quiescent infection appears to be a fungal response to adverse physiological conditions temporarily imposed by the host. . The lack of available nutrients, the presence of preformed antifungal compounds and the lack of enzymatic potential to penetrate have been tested as possible causes of quiescent fungal infection on unripe fruits (Pruskyel aI., 1998).

Binyamini & Schiffmanri-Nadel (1972) suggested that the lack of development of C.

gloeosporioides on avocado fruits while still on the tree might be because the hard exterior of the fruit contains fungal growth inhibitors. Verhoeff (1974) and Prusky et al.

(1982) reported a preformed .anti-fungal compound, l-acetoxy-2-hydroxy-4-oxo-heneicosa-12, 15-diene (diene), associated with the resistance of fruits to attack by post-harvest pathogens. In freshly harvested fruits, the concentration of antifungal diene drops to subfungitoxic concentrations and then increases after a lag period to the senescence of leaves, ripening of fruits, or wounding. Senescing leaves have almost completed their role in supplying photosynthate and are therefore of little further economic value but fruits increase in value during ripening. It is perhaps for this reason that latency has been studied more in fruits than in leaves (Muirhead, 1981).

In Colletotrichum species, the role of appressoria and infection hyphae as latent structures has often been debated and two forms of latency have been reported. Appressoria remain dormant on the surface of the fruit, giving rise to penetration hyphae and infection hyphae only at ripening (Binyamini & Schiffmann-Nadel, 1972; Parbery & Emmett, 1977). Another form of latency is when appressoria germinate immediately, penetrate, and produce a few latent subcuticular hyphae, which may resume activity during ripening (Verhoeff, 1974; Chau & Alvarez, 1983; Coates et al., 1993; Prusky et .

concentrations present in pre-harvested fruits (Karni et al., 1989; Prusky ef al., 1991). The process leading to the fast decrease of the antifungal diene was suggested to be similar to the process occurring in normal ripening fruits, where the breakdown of the anti-fungal diene is catalyzed by lipoxygenase whose activity is regulated by the flavan 3-01 epicatechin (Prusky et al., 1985, 1988). In freshly harvested fruit, a fast increase in epicatechin was observed, accompanied by an increase in lipoxygenase activity and consequently a decrease in the concentration of the anti-fungal diene (Karni et al., 1989). Infiltration of fruits with a-tocopherol, an inhibitor of lipoxygenase activity, inhibited the decrease of antifungal diene and also the development of anthracnose lesions (Prusky

et al., 1983). The induction of epicatechin enhanced the level of the antifunfal compounds by preventing its breakdown. Induced levels of epicatechin occur by activation of the phenylpropanoid pathway with enhanced activity of PAL, CHS, and flavanone 3 hydroxylase. Glomerela cingula/a (Stonem.) Spauld. & Schrenk also decays the mesocarp of peeled unripe fruits despite fungitoxic concentrations of the antifungal diene in idioblast oil cells. These idioblasts are metabolically active and can incorporate labeled precursors and synthesise and export diene (Prusky e/ al., 1998).

Reactive oxygen species (ROS) induced by fungal infection of unripe fruits may modulate resistance, resulting in the inhibition of fungal development and quiescence. Both artificial inoculation of avocado pericarp tissue with C. gloeosporioides (G.

cingulata) and treatment of avocado cell cultures with the cell wall elicitor of G.

cingula/a. increased the production of ROS, like hydrogen peroxide (H202). Unripe, resistant fruits are physiologically able to produce twice levels of ROS in pericarp tissue compared to susceptible tissue. Exogenous application of H202 at a rate of 1 mM to

pericarp tissue enhanced ROS, phenylalanine ammonia lyase (PAL) activity, and epicatechin levels (Beno-Moualern &Prusky, 2000).

Treatment of host tissue after harvest with chemicals may cause nutrients to be released or made more available at the fruit surface. Ethylene increased or reduced the incidence of disease depending upon the time and concentration of the ethylene treatment and artificial inoculation. Upon exposure of inoculated green colored tangerines to ethylene, the infection process and resistance mechanisms were apparently initiated simultaneously, but despite this the infection process develops more rapidly (Brown & Barmore, 1977). The infection of tomato fruit by

C.

gloeosporioides did not

proceed until the onset of ripening in response to ethylene. Compared with fruit from wild-type plants, infection progressed more slowly in transgenie fruit in which ethylene biosynthesis and ripening had been inhibited by an ACC oxidase (ACO) anti sense transgene (Cooper et al., 1998). Furthermore, l-aminocylopropane l-carboxylic acid oxidase I(ACOl) mRNA, an enzyme responsible for the conversion of S-adenosyl methionine to ethylene, accumulated to maximum levels during the early stages of infection in ripe and unripe fruit. Ethylene biosynthesis increased rapidly in response to infection of ripe wild type and l-aminocylopropane I-carboxylic acid oxidase (ACO) anti sense fruit but was 25% times greater in the former. Furthermore, ripening fruit from the mutant ripening inhibitor (rin), which are normally very resistant to infection, became infected quickly when incubated in the presence of ethylene, whereas fruit incubated in the absence of ethylene remained healthy.

Ethylene could possibly alter the quantity or quality of cellulolytic or pectolytic enzymes produced by

C.

gloeosporioides, which subsequently could affect its ability to colonize host tissue (Brown, 1975; Brown & Barmore, 1977). Transgenie tomato fruit deficient in polygalacturonase developed lesions at the same rate as the wild fruit type did to

C.

gloeosporioides infection. This suggests that depolymerization of pectic substances has little bearing on the progress of post harvest disease (Cooper et al., 1998)

1.3.4 Infection strategies

Colletotricum species exhibit two main infection strategies, according to which

species are loosely categorized. Skipp et al., (1995) and Bailey et al., (1992) have described the infection strategies employed by Colletotricum species and the reader is specifically referred to a review by Bailey et al., (1992) for the list of Colletotrichum species and their respective infection strategies. Recent molecular aspects of these infection strategies will be dealt here vis-a-vis the aspects discussed in the aforementioned reviews. The first groups of species are known as intracellular hemibiotrophs; they penetrate both the cuticle and the epidermal cell wall and then invade the tissues. Some of the species in this group do not have a recognized biotrophic phase (Mould et al., 1991a, b). In most cases, this group displays a two-phase infection process whereby the sequence of transient biotrophic phase followed by slow senescence and eventual death of the infected cells is repeated (Bailey et al., 1992; Skipp

et al., 1995; Morin et al., 1996). The pathogenicity and possibly the host specificity of

hemibiotrophic Colletotrichum species may be determined during the early, apparently

biotrophic stage of infection (Bailey et al., 1992; O'Connell, et al., 1993). The visible symptoms caused by Colletotrichum species are primarily a result of the later stage of necrotrophic development (Wei et al., 1997).

During the first phase, the pathogen grows biotrophically and no symptom IS

produced on the plant. Immediately after penetration, infection vesicles grow and fungal hyphae then grow between the plasma membrane and plant cell wall. These vesicles give rise to primary hyphae, which, ~lso branch and grow rapidly until the initially infected cells are packed with convoluted mycelium (O'Connell et al., 1985; O'Connell et al., 1993; Skipp et al., 1995; Latunde-Dada et al., 1997; Smith et al., 1999) or spread in to several nearby epidermal host cells (O'Connell et al., 1985;